Can AI help clear the COVID-19 breast screening backlog?

By Dr. Christopher Austin, AuntMinnie.com contributing writer
Originally published on AuntMinnie

As in nearly every aspect of our lives, the impact of COVID-19 on the world of breast cancer screening has been unprecedented, with worrisome short- and long-term consequences. In fact, a recently published study in JAMA Oncology1 revealed that 3.9 million women in the U.S. missed their routine screening mammograms due to the pandemic, and cancer charity Breast Cancer Now found that nearly 1 million women in the U.K. did as well.2 This has resulted in an ongoing backlog of patients as breast screening clinics have reopened with reduced capacity due to increased COVID-19 safety protocols.

Breast cancer represents the most common cancer among women and accounts for an estimated 500,000 deaths per year worldwide.3 Population-based screening remains a proven service for substantially reducing breast cancer mortality; that is, until it was compromised by the pandemic.

The number of potentially undiagnosed breast cancers due to COVID-related delays is troubling. Even more problematic is the question of whether breast screening providers will be able to expedite and prioritize the screening process for the highest-risk women who are overdue for their regular mammograms.

Even as public health conditions have improved, breast screening services are still not operating at pre-pandemic capacity, which further exacerbates the backlog created by the COVID-19 pandemic. Unfortunately, the rate of undetected breast cancers and resulting deaths is expected to increase due to delayed screening.4

In addition to direct COVID-19 related delays, breast screening centers have experienced increased recalls due to the incidence of lymphadenopathy on screening mammograms, largely attributed to recent COVID-19 vaccine uptake. Vaccine-induced lymphatic swelling has been shown to result in a rise of false-positive screening cases during routine mammography.5 This not only causes unwarranted anxiety for the women who are recalled, but also adds to growing patient backlogs.

In the wake of the COVID-19 pandemic, the question remains as to how breast screening providers can identify and prioritize breast screening for women who face the most risk from having their screening appointments delayed. To find answers to this question, we can take a closer look at issues affecting the breast screening community, an overview of the impact of COVID-19 on the breast cancer pathway, and how new developments in artificial intelligence (AI) can offer solutions.

The impact of COVID-19 on the breast cancer screening pathway

Even prior to the COVID-19 pandemic, breast screening programs faced numerous challenges. Burnout and fatigue among radiologists represent significant hurdles for breast screening clinics, exacerbated by an overall shortage of qualified radiology professionals.

From screening and prevention to diagnosis and treatment, COVID-19 has created a multitude of new issues for providers and patients throughout the breast cancer screening pathway in their ability to detect breast cancer and treat it expeditiously.

Recent studies6,7 have shown that COVID-19 affected access to breast screening services by deepening existing socioeconomic disparities among minorities and rural populations. In addition, at the point of image acquisition, the COVID-19 pandemic has reduced screening capacity due to public health safety measures and reduced patient participation in routine screening.

Avoiding delays to a screening is also of the essence. It is estimated that a screening delay of six months may result in an additional two breast cancer deaths per thousand.8 Furthermore, new research indicates pandemic-related diagnosis delays may lead to an estimated 7.9%-9.6% increase in the number of deaths due to breast cancer up to five years after diagnosis.9

At each stage of the breast screening pathway, COVID-19 presented obstacles. Public health lockdowns reduced access to care with the closures of screening centers, yet once they reopened, many women further delayed their screenings due to fear of infection. Screening centers have experienced new challenges in administering care with social distancing requirements and increased time and resources required for disinfection. In addition, some centers had to defer routine diagnostic work that normally would allow breast cancer to be diagnosed and treated earlier.10

These delays may collectively contribute to later-stage diagnosis, during which women have a lower probability of surviving breast cancer.11,12 Those who already were diagnosed when the COVID-19 pandemic began faced the reality of delayed, modified, or postponed treatment.

As breast screening clinics resume operations, they face ongoing scheduling challenges in addressing the backlog of delayed mammograms, which are magnified by reduced imaging staff availability and increased safety protocols. According to some experts, the backlogs will continue to grow despite clinic reopenings due to reduced capacity.13

The issue of burnout and fatigue among radiologists and radiologic technologists/radiographers has also persisted and likely worsened under the strain of the pandemic. This negatively impacts image acquisition capacity and impacts workforce morale. If these issues are not managed, each of these factors can lead to a vicious cycle in which breast screening backlogs are extended rather than curbed.

How can breast screening centers catch up?

As a first step to breaking the cycle, clinics can accelerate their adoption of artificial intelligence (AI) throughout the breast-screening pathway. This starts with prioritizing the list of unscreened women, inviting those with an identified elevated cancer risk first, and then working through the remaining patient list as efficiently as possible. This approach aims to reduce any further delays that could lead to worse outcomes for those women most at risk -- including potentially lower rates of survival and higher treatment costs.

Coming out of the pandemic, breast screening clinics do not have the tools to perform prioritized scheduling on their screening backlog. AI, however, is opening new pathways to directly address this challenge, with AI-enabled algorithms that can help identify those women who should be scheduled first for their screening based on an unbiased scoring of their potential to develop breast cancer.

Kheiron Medical Technologies, an award-winning applied science company based in the U.K., is launching RSViP™, a new AI-enabled tool that helps breast screening programs to optimize scheduling and screening for women who are at highest risk for developing breast cancer. It enables radiologists and breast screening programs to move the highest-risk women to the front of the line, which means earlier diagnosis, a better chance of survival, lower healthcare costs, and reduced anxiety for both the patient and provider.

RSViP works by analyzing images from each woman's prior screening, using machine-learning insights to provide a score based on the potential for developing breast cancer. Its core algorithm comes from Mia®, Kheiron's comprehensive suite of AI-enabled solutions for the entire breast-screening pathway, which is proven to perform at the level of a radiologist on a representative screening population. Based on an unbiased assessment for every unscreened woman, RSViP generates an actionable scheduling list that the breast screening program can use to schedule appointments for women who are considered to be most at risk of developing breast cancer.

The journey through the COVID-19 pandemic has been longer than anyone would have expected, and there remains a long road ahead to recover and unbury the breast screening community from its current backlog. However, there is hope on the horizon with new approaches and technology applications powered by artificial intelligence.

It only requires an open mind to embrace new technology and processes to help clear the backlog to achieve what is most important -- helping to save lives in the fight against breast cancer.


References

  1. Chen RC, Haynes K, Du S, Barron J, Katz AJ. Association of Cancer Screening Deficit in the United States With the COVID-19 Pandemic. JAMA Oncol. 2021;7(6):878-884.
  2. Almost one million women in UK miss vital breast screening due to COVID-19. BreastCancerNow. https://www.breastcancernow.org/about-us/media/press-releases/almost-one-million-women-in-uk-miss-vital-breast-screening-due-covid-19.
  3. FFerlay J, Soerjomataram I, Dikshit R, Eser S, Mathers C, Rebelo M, Parkin DM, Forman D, Bray F. Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012. Int J Cancer. 2015 Mar 1;136(5):E359-86.
  4. Brewster A. How to Safely Keep Up With Your Cancer Screening and Care During COVID-19. Cancer.net. https://www.cancer.net/blog/2020-12/how-safely-keep-with-your-cancer-screening-and-care-during-covid-19. December 10, 2020.
  5. COVID-19 Vaccine: Can It Affect Your Mammogram Results? John Hopkins Medicine website. https://www.hopkinsmedicine.org/health/conditions-and-diseases/coronavirus/covid19-vaccine-can-it-affect-your-mammogram-results.
  6. Allegretto A. COVID-19 widens disparities in breast screening. AuntMinnie.com. May 24, 2021. https://www.auntminnie.com/index.aspx?sec=sup&sub=wom&pag=dis&ItemID=132478.
  7. Amram O, Robison J, Amiri S, Pflugeisen B, Roll J, Monsivais P. Socioeconomic and Racial Inequities in Breast Cancer Screening During the COVID-19 Pandemic in Washington State. JAMA Netw Open. 2021 May 3;4(5):e2110946.
  8. Kregting L, Kaljouw S, de Jonge L, Jansen E, Peterse E, Heijnsdijk E, et al. Effects of cancer screening restart strategies after COVID-19 disruption. European Journal of Cancer. 2020 Oct;138:S16.
  9. Maringe, C., Spicer, J., Morris, M., Purushotham, A., Nolte, E., Sullivan, R., Rachet, B., & Aggarwal, A. (2020). The impact of the COVID-19 pandemic on cancer deaths due to delays in diagnosis in England, UK: a national, population-based, modelling study. Lancet Oncol. 2020;21(8), 1023-1034.
  10. Ibid
  11. Epic Health Research Network. Preventive cancer screenings during COVID-19 pandemic. May 1, 2020. Accessed August 20, 2020. https://www.ehrn.org/wp-content/ uploads/Preventive-Cancer-Screenings- during-COVID-19-Pandemic.pdf
  12. Maringe C, Spicer J, Morris M, et al. The impact of the COVID-19 pandemic on cancer deaths due to delays in diagnosis in England, UK: a national, population-based, modelling study. Lancet Oncol. 2020;21(8):1023-1034.
  13. How COVID-19 Created a Health Care Backlog. Knowledge@Wharton. https://knowledge.wharton.upenn.edu/article/how-covid-19-created-a-health-care-backlog/.